Strawflower…And More
The image above is an artistic rendering of a beautiful Strawflower that Bonnie recently planted in a planter in our backyard. The image directly below is a photograph of that same plant and its blooms. The difference between the two images is due to what I did and what I didn’t instruct the AI to do. For something as intricate as a flower, I either can’t describe it fully or lack the vocabulary, but I can add style and imagination that aren’t present in the original photo.
Many may prefer the original photo (directly below), and I can understand that, as it would be my preference in many cases as well. At least some preference is determined by the final image’s purpose or presentation.
.
For the final image below, I instructed the AI to create a depiction of a strawflower, and while I described the settings and circumstances under which the image should be produced, much like one would specify the details for a photograph, I did not provide a description of the strawflower itself, as I assumed that the AI already knows what a strawflower looks like or can easily find that information.
For your reference, I have included the complete instructions (prompt) I used below the AI-generated image from the prompt. Interestingly, this same prompt can produce a unique image each time it is used, even with the same AI, and especially when using different AIs.
AI Prompt: “Studio product macro, ultra-detailed ā near-micro image of a yellow fully open strawflower bloom with buds and foliage rendered in photographic commercial quality. Multi-shot focus stacking for edge-to-edge sharpness, captured with a high-res medium-format sensor and a 120mm macro at f/8, ISO 50, with controlled strobe lighting (soft key + hair light) to create a gentle wrap and a crisp rim. 45-degree angled composition, slight tilt for dynamic flow, glossy dew accent, color-corrected RAW, retouched with clarity and micro-contrast to reveal every fiber of the strawflowerās papery petals and bud textures for print and advertising use.”
I like all three renderings, Earl. Conventional photography would have a difficult time trying to reproduce results like this, especially considering the amount of time it takes to make these. I haven’t been to school in a long while, but I’m willing to bet a week’s pay that they are teaching photography students AI reproductions at RIT. I haven’t jumped into the pool yet but it’s only a matter of time. Good work!
It’s a bit like the wild west with AI graphics/art at the moment, but that’s been the norm for most new media and methods. At this point, I’m finding that it still takes a lot of human interaction/skill/imagination to take an image beyond the minimum to make it into what might be called art. Thanks, Ken, and it’s good to hear from you. Have a good weekend!
I am amazed how AI and you are creating such art! My favorite is the third one. And, that may be because it does not look like a photograph. Show me more,please.
I’m sure there will be more examples of photography and AI here. The more I explore, the more possibilities present themselves.
Very interesting and quite beautiful result. The AI image has a clear three dimensionality that really makes it pop off the page.
Thanks, Howard. I find it fascinating not only because it raises questions about the process, the image, and art, but also for other reasons. There is a sense of freedom involved. If you can imagine something and describe it in enough detail, AI can likely create it fairly well. Iām not at that level yet, and there’s a good chance I’ll hit my own limits before I discover any with AI graphics. The old brain is not as flexible as it used to be. š¤